4 .. **Read this first.**
6 The text below explains how to update the release notes.
8 Use proper spelling, capitalization and punctuation in all sections.
10 Variable and config names should be quoted as fixed width text: ``LIKE_THIS``.
12 Build the docs and view the output file to ensure the changes are correct::
16 firefox build/doc/html/guides/rel_notes/release_16_11.html
22 .. This section should contain new features added in this release. Sample format:
24 * **Add a title in the past tense with a full stop.**
26 Add a short 1-2 sentence description in the past tense. The description
27 should be enough to allow someone scanning the release notes to understand
30 If the feature adds a lot of sub-features you can use a bullet list like this.
32 * Added feature foo to do something.
33 * Enhanced feature bar to do something else.
35 Refer to the previous release notes for examples.
37 This section is a comment. Make sure to start the actual text at the margin.
40 * **Added software parser for packet type.**
42 * Added a new function ``rte_pktmbuf_read()`` to read the packet data from an
43 mbuf chain, linearizing if required.
44 * Added a new function ``rte_net_get_ptype()`` to parse an Ethernet packet
45 in an mbuf chain and retrieve its packet type by software.
46 * Added new functions ``rte_get_ptype_*()`` to dump a packet type as a string.
48 * **Improved offloads support in mbuf.**
50 * Added a new function ``rte_raw_cksum_mbuf()`` to process the checksum of
51 data embedded in an mbuf chain.
52 * Added new Rx checksum flags in mbufs to describe more states: unknown,
53 good, bad, or not present (useful for virtual drivers). This modification
54 was done for IP and L4.
55 * Added a new RX LRO mbuf flag, used when packets are coalesced. This
56 flag indicates that the segment size of original packets is known.
58 * **Added vhost-user dequeue zero copy support**
60 The copy in dequeue path is saved, which is meant to improve the performance.
61 In the VM2VM case, the boost is quite impressive. The bigger the packet size,
62 the bigger performance boost you may get. However, for VM2NIC case, there
63 are some limitations, yet the boost is not that impressive as VM2VM case.
64 It may even drop quite a bit for small packets.
66 For such reason, this feature is disabled by default. It can be enabled when
67 ``RTE_VHOST_USER_DEQUEUE_ZERO_COPY`` flag is given. Check the vhost section
68 at programming guide for more information.
70 * **Added vhost-user indirect descriptors support.**
72 If indirect descriptor feature is negotiated, each packet sent by the guest
73 will take exactly one slot in the enqueue virtqueue. Without the feature, in
74 current version, even 64 bytes packets take two slots with Virtio PMD on guest
77 The main impact is better performance for 0% packet loss use-cases, as it
78 behaves as if the virtqueue size was enlarged, so more packets can be buffered
79 in case of system perturbations. On the downside, small performance degradation
80 is measured when running micro-benchmarks.
82 * **Added vhost PMD xstats.**
84 Added extended statistics to vhost PMD from per port perspective.
86 * **Supported offloads with virtio.**
92 * **Added virtio NEON support for ARM.**
94 * **Updated the ixgbe base driver.**
96 Updated the ixgbe base driver, including the following changes:
98 * add X550em_a 10G PHY support
99 * support flow control auto negotiation for X550em_a 1G PHY
100 * add X550em_a FW ALEF support
101 * increase mailbox version to ixgbe_mbox_api_13
102 * add two MAC ops for Hyper-V support
104 * **Added API's for VF management to the ixgbe PMD.**
106 Eight new API's have been added to the ixgbe PMD for VF management from the PF.
107 The declarations for the API's can be found in ``rte_pmd_ixgbe.h``.
109 * **Updated the enic driver.**
111 * Use interrupt for link status checking instead of polling
112 * More flow director modes on UCS Blade with firmware version >= 2.0(13e)
113 * Full support for MTU update
114 * Support for rte_eth_rx_queue_count function
116 * **Updated the mlx5 driver.**
118 * Add support for RSS hash result
119 * Several performance improvements
122 * **Updated the QAT PMD.**
124 The QAT PMD was updated with following support:
127 * SHA224-HMAC algorithm
128 * SHA384-HMAC algorithm
130 * KASUMI (F8 and F9) algorithm
136 * **Added openssl PMD.**
138 A new crypto PMD has been added, which provides several ciphering and hashing.
139 All cryptography operations are using Openssl library crypto API.
141 * **Updated the IPsec example with following support:**
144 * AES CBC IV generation with cipher forward function
147 * **Added support for new gcc -march option.**
149 The GCC 4.9 ``-march`` option supports the Intel processor code names.
150 The config option ``RTE_MACHINE`` can be used to pass code names to the compiler as ``-march`` flag.
156 .. This section should contain bug fixes added to the relevant sections. Sample format:
158 * **code/section Fixed issue in the past tense with a full stop.**
160 Add a short 1-2 sentence description of the resolved issue in the past tense.
161 The title should contain the code/lib section like a commit message.
162 Add the entries in alphabetic order in the relevant sections below.
164 This section is a comment. Make sure to start the actual text at the margin.
174 * **enic: Fixed several flow director issues.**
176 * **enic: Fixed inadvertent setting of L4 checksum ptype on ICMP packets.**
178 * **enic: Fixed high driver overhead when servicing Rx queues beyond the first.**
196 .. This section should contain new known issues in this release. Sample format:
198 * **Add title in present tense with full stop.**
200 Add a short 1-2 sentence description of the known issue in the present
201 tense. Add information on any known workarounds.
203 This section is a comment. Make sure to start the actual text at the margin.
205 * **L3fwd-power app does not work properly when Rx vector is enabled.**
207 Using some drivers with vector enabled, makes L3fwd-power app not work
208 properly, since the queue monitoring works differently when using
209 scalar compared to vector, making the frequency scaling not to work correctly.
210 In addition, L3fwd-power requires the mbuf to have correct packet type,
211 but in some drivers, the vector mode must be disabled for this.
213 Therefore, in order to use L3fwd-power, vector mode should be disabled
214 from the config file.
216 * **Digest address must be supplied for crypto auth operation on QAT PMD.**
218 The cryptodev API specifies that if the rte_crypto_sym_op.digest.data field,
219 and by inference the digest.phys_addr field which points to the same location,
220 is not set for an auth operation the driver is to understand that the digest
221 result is located immediately following the region over which the digest is
222 computed. The QAT PMD doesn't correctly handle this case and reads and writes
223 to an incorrect location.
225 Callers can workaround this by always supplying the digest virtual and
226 physical address fields in the rte_crypto_sym_op for an auth operation.
232 .. This section should contain API changes. Sample format:
234 * Add a short 1-2 sentence description of the API change. Use fixed width
235 quotes for ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
237 This section is a comment. Make sure to start the actual text at the margin.
239 * The driver names have been changed. It especially impacts ``--vdev`` arguments.
240 Examples: ``eth_pcap`` becomes ``net_pcap``
241 and ``cryptodev_aesni_mb_pmd`` becomes ``crypto_aesni_mb``.
243 * The log history is removed.
245 * The ``rte_ivshmem`` feature (including library and EAL code) has been removed
246 in 16.11 because it had some design issues which were not planned to be fixed.
248 * The ``file_name`` data type of ``struct rte_port_source_params`` and
249 ``struct rte_port_sink_params`` is changed from `char *`` to ``const char *``.
251 * **Improved device/driver hierarchy and generalized hotplugging**
253 Device and driver relationship has been restructured by introducing generic
254 classes. This paves way for having PCI, VDEV and other device types as
255 just instantiated objects rather than classes in themselves. Hotplugging too
256 has been generalized into EAL so that ethernet or crypto devices can use the
257 common infrastructure.
259 * removed ``pmd_type`` as way of segregation of devices
260 * moved ``numa_node`` and ``devargs`` into ``rte_driver`` from
261 ``rte_pci_driver``. These can now be used by any instantiated object of
263 * added ``rte_device`` class and all PCI and VDEV devices inherit from it
264 * renamed devinit/devuninit handlers to probe/remove to make it more
265 semantically correct with respect to device<=>driver relationship
266 * moved hotplugging support to EAL. Hereafter, PCI and vdev can use the
267 APIs ``rte_eal_dev_attach`` and ``rte_eal_dev_detach``.
268 * helpers and support macros have been renamed to make them more synonymous
269 with their device types
270 (e.g. ``PMD_REGISTER_DRIVER`` => ``RTE_PMD_REGISTER_PCI``)
271 * Device naming functions have been generalized from ethdev and cryptodev
272 to EAL. ``rte_eal_pci_device_name`` has been introduced for obtaining
273 unique device name from PCI Domain-BDF description.
274 * Virtual device registration APIs have been added: ``rte_eal_vdrv_register``
275 and ``rte_eal_vdrv_unregister``.
281 .. This section should contain ABI changes. Sample format:
283 * Add a short 1-2 sentence description of the ABI change that was announced in
284 the previous releases and made in this release. Use fixed width quotes for
285 ``rte_function_names`` or ``rte_struct_names``. Use the past tense.
287 This section is a comment. Make sure to start the actual text at the margin.
291 Shared Library Versions
292 -----------------------
294 .. Update any library version updated in this release and prepend with a ``+``
299 + librte_cfgfile.so.2
304 The libraries prepended with a plus sign were incremented in this version.
311 + librte_cryptodev.so.2
312 librte_distributor.so.1
342 .. This section should contain a list of platforms that were tested with this release.
351 This section is a comment. Make sure to start the actual text at the margin.
353 #. Intel(R) Server board S2600WTT
355 - Processor: Intel(R) Xeon(R) CPU E5-2697 v2 @ 2.70GHz
359 - Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
363 - Intel(R) Xeon(R) CPU E5-2697 v3 @ 2.60GHz
367 - Machine type-model: 8247-22L
368 - Firmware FW810.21 (SV810_108)
369 - Processor: POWER8E (raw), AltiVec supported
375 .. This section should contain a list of NICs that were tested with this release.
384 This section is a comment. Make sure to start the actual text at the margin.
386 #. Mellanox(R) ConnectX(R)-4 10G MCX4111A-XCAT (1x10G)
388 * Host interface: PCI Express 3.0 x8
389 * Device ID: 15b3:1013
390 * MLNX_OFED: 3.4-1.0.0.0
391 * Firmware version: 12.17.1010
393 #. Mellanox(R) ConnectX(R)-4 10G MCX4121A-XCAT (2x10G)
395 * Host interface: PCI Express 3.0 x8
396 * Device ID: 15b3:1013
397 * MLNX_OFED: 3.4-1.0.0.0
398 * Firmware version: 12.17.1010
400 #. Mellanox(R) ConnectX(R)-4 25G MCX4111A-ACAT (1x25G)
402 * Host interface: PCI Express 3.0 x8
403 * Device ID: 15b3:1013
404 * MLNX_OFED: 3.4-1.0.0.0
405 * Firmware version: 12.17.1010
407 #. Mellanox(R) ConnectX(R)-4 25G MCX4121A-ACAT (2x25G)
409 * Host interface: PCI Express 3.0 x8
410 * Device ID: 15b3:1013
411 * MLNX_OFED: 3.4-1.0.0.0
412 * Firmware version: 12.17.1010
414 #. Mellanox(R) ConnectX(R)-4 40G MCX4131A-BCAT/MCX413A-BCAT (1x40G)
416 * Host interface: PCI Express 3.0 x8
417 * Device ID: 15b3:1013
418 * MLNX_OFED: 3.4-1.0.0.0
419 * Firmware version: 12.17.1010
421 #. Mellanox(R) ConnectX(R)-4 40G MCX415A-BCAT (1x40G)
423 * Host interface: PCI Express 3.0 x16
424 * Device ID: 15b3:1013
425 * MLNX_OFED: 3.4-1.0.0.0
426 * Firmware version: 12.17.1010
428 #. Mellanox(R) ConnectX(R)-4 50G MCX4131A-GCAT/MCX413A-GCAT (1x50G)
430 * Host interface: PCI Express 3.0 x8
431 * Device ID: 15b3:1013
432 * MLNX_OFED: 3.4-1.0.0.0
433 * Firmware version: 12.17.1010
435 #. Mellanox(R) ConnectX(R)-4 50G MCX414A-BCAT (2x50G)
437 * Host interface: PCI Express 3.0 x8
438 * Device ID: 15b3:1013
439 * MLNX_OFED: 3.4-1.0.0.0
440 * Firmware version: 12.17.1010
442 #. Mellanox(R) ConnectX(R)-4 50G MCX415A-GCAT/MCX416A-BCAT/MCX416A-GCAT (2x50G)
444 * Host interface: PCI Express 3.0 x16
445 * Device ID: 15b3:1013
446 * MLNX_OFED: 3.4-1.0.0.0
447 * Firmware version: 12.17.1010
449 #. Mellanox(R) ConnectX(R)-4 50G MCX415A-CCAT (1x100G)
451 * Host interface: PCI Express 3.0 x16
452 * Device ID: 15b3:1013
453 * MLNX_OFED: 3.4-1.0.0.0
454 * Firmware version: 12.17.1010
456 #. Mellanox(R) ConnectX(R)-4 100G MCX416A-CCAT (2x100G)
458 * Host interface: PCI Express 3.0 x16
459 * Device ID: 15b3:1013
460 * MLNX_OFED: 3.4-1.0.0.0
461 * Firmware version: 12.17.1010
463 #. Mellanox(R) ConnectX(R)-4 Lx 10G MCX4121A-XCAT (2x10G)
465 * Host interface: PCI Express 3.0 x8
466 * Device ID: 15b3:1015
467 * MLNX_OFED: 3.4-1.0.0.0
468 * Firmware version: 14.17.1010
470 #. Mellanox(R) ConnectX(R)-4 Lx 25G MCX4121A-ACAT (2x25G)
472 * Host interface: PCI Express 3.0 x8
473 * Device ID: 15b3:1015
474 * MLNX_OFED: 3.4-1.0.0.0
475 * Firmware version: 14.17.1010
481 .. This section should contain a list of OSes that were tested with this release.
482 The format is as follows, in alphabetical order:
488 * Red Hat Enterprise Linux 7.2
489 * SUSE Enterprise Linux 12
494 This section is a comment. Make sure to start the actual text at the margin.
496 * Red Hat Enterprise Linux Server release 6.7 (Santiago)
497 * Red Hat Enterprise Linux Server release 7.0 (Maipo)
498 * Red Hat Enterprise Linux Server release 7.2 (Maipo)
499 * Wind River Linux 6.0.0.26